Learning Decomposed Representation for Counterfactual Inference

TL;DR

This paper introduces a novel framework that decomposes observed variables to better identify confounders, leading to more accurate treatment effect estimation from observational data.

Contribution

It proposes a synergistic learning approach that decomposes representations to distinguish confounders from non-confounders, improving causal inference accuracy.

Findings

More precise confounder decomposition on synthetic data

Enhanced treatment effect estimation on real-world datasets

Outperforms baseline methods in accuracy

Abstract

The fundamental problem in treatment effect estimation from observational data is confounder identification and balancing. Most of the previous methods realized confounder balancing by treating all observed pre-treatment variables as confounders, ignoring further identifying confounders and non-confounders. In general, not all the observed pre-treatment variables are confounders that refer to the common causes of the treatment and the outcome, some variables only contribute to the treatment and some only contribute to the outcome. Balancing those non-confounders, including instrumental variables and adjustment variables, would generate additional bias for treatment effect estimation. By modeling the different causal relations among observed pre-treatment variables, treatment and outcome, we propose a synergistic learning framework to 1) identify confounders by learning decomposed representations of both confounders and non-confounders, 2) balance confounder with sample re-weighting technique, and simultaneously 3) estimate the treatment effect in observational studies via counterfactual inference. Empirical results on synthetic and real-world datasets demonstrate that the proposed method can precisely decompose confounders and achieve a more precise estimation of treatment effect than baselines.